A1058
Title: A multivariate compound dynamic contagion process for infectious events
Authors: Rosy Oh - Korea Military Academy (Korea, South) [presenting]
Abstract: The dynamic contagion process (DCP) is a generalization of the externally exciting Cox process with shot noise intensity and the self-exciting Hawkes process. To date, however, no theoretical work has been done on a multivariate dynamic contagion process (MDCP) and its compound process. The theoretical distributional properties for these processes are analyzed systematically, based on the piecewise deterministic Markov process theory. Exact simulation algorithms are provided for these processes. Numerical results show that these processes can be used for the modelling of aggregate loss and the number of infections arising from contagious catastrophic events.
